Amec Foster Wheeler Project No. 6064150003 1 Amec Foster Wheeler Environment & Infrastructure, Inc. CASTNET Summary of Quarterly Operations (January through March) EPA Contract No. EP-W-15-003 Introduction This quarterly report summarizes results from the Clean Air Status and Trends Network (CASTNET) quality assurance/quality control (QA/QC) program for data collected during first quarter 2016. The various QA/QC criteria and policies are documented in the CASTNET Quality Assurance Project Plan (QAPP; Amec Foster Wheeler, 2014). The QAPP is comprehensive and includes standards and policies for all components of project operation from site selection through final data reporting. It is reviewed annually and updated as warranted. Quarterly Summary The annual management review meeting in support of maintaining International Organization for Standardization (ISO)/International Electrotechnical Commission (IEC) 17025:2005 accreditation by the American Association for Laboratory Accreditation (A2LA) was held in early January 2016. The meeting participants discussed the annual management review report distributed to the management group in December 2015. The agenda covered discussion of the status of the program as well as program goals for 2015 and 2016. During review of third quarter 2015 filter pack sulfur dioxide (SO 2 ) concentrations in January 2016, as part of the preparation of the CASTNET Third Quarter 2015 Data Report, it became apparent that aggregates of the SO 2 data for the eastern reference sites were not reasonable when compared with long-term third quarter time series. Amec Foster Wheeler immediately began evaluating possible causes of the anomalous measurements. The EPA Contract-level Project Officer was notified as per ISO 17025 management system protocol. After this discussion, Amec Foster Wheeler removed the SO 2 -related discussions from the third quarter report and worked with EPA to remove these data from the publicly available database. The cause of the anomalous measurements correlates with the laboratory’s change to a different supplier of the reagent used to prepare the cellulose filter impregnation solution. This supplier was used for filter pack samples that ran from mid-May through December 2015. The prepared impregnation solutions passed acceptance testing prior to use. The cellulose filter extracts were tested for elevated
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
CASTNET Summary of Quarterly Operations (January through March) EPA Contract No. EP-W-15-003
Introduction This quarterly report summarizes results from the Clean Air Status and Trends Network (CASTNET) quality assurance/quality control (QA/QC) program for data collected during first quarter 2016. The various QA/QC criteria and policies are documented in the CASTNET Quality Assurance Project Plan (QAPP; Amec Foster Wheeler, 2014). The QAPP is comprehensive and includes standards and policies for all components of project operation from site selection through final data reporting. It is reviewed annually and updated as warranted. Quarterly Summary The annual management review meeting in support of maintaining International Organization for Standardization (ISO)/International Electrotechnical Commission (IEC) 17025:2005 accreditation by the American Association for Laboratory Accreditation (A2LA) was held in early January 2016. The meeting participants discussed the annual management review report distributed to the management group in December 2015. The agenda covered discussion of the status of the program as well as program goals for 2015 and 2016. During review of third quarter 2015 filter pack sulfur dioxide (SO
2) concentrations in January 2016, as part of the preparation of the CASTNET Third Quarter 2015 Data Report, it became apparent that aggregates of the SO
2 data for the eastern reference sites were not reasonable when compared with long-term third quarter time series. Amec Foster Wheeler immediately began evaluating possible causes of the anomalous measurements. The EPA Contract-level Project Officer was notified as per ISO 17025 management system protocol. After this discussion, Amec Foster Wheeler removed the SO
2-related discussions from the third quarter report and worked with EPA to remove these data from the publicly available database. The cause of the anomalous measurements correlates with the laboratory’s change to a different supplier of the reagent used to prepare the cellulose filter impregnation solution. This supplier was used for filter pack samples that ran from mid-May through December 2015. The prepared impregnation solutions passed acceptance testing prior to use. The cellulose filter extracts were tested for elevated
nitrate (NO- 3) concentrations as well as sulfate (SO2-
4) concentrations to see if something was enhancing conductivity responses for detected analytes. Amec Foster Wheeler’s laboratory re-analyzed previously exposed samples from selected eastern and northeastern CASTNET sites. Samples from June through September show higher NO-
3, which was not seen in previously exposed samples that used reagent from the usual supplier. Amec Foster Wheeler began running statistical analyses for further evaluation of the Teflon and nylon filter SO2-
4 concentrations. While some of these tests seem to indicate a suspect reagent, they are not conclusive. The investigation to determine root cause is, therefore, ongoing. Amec Foster Wheeler is also evaluating additional acceptance procedures for new suppliers and screening procedures to facilitate early identification of results that are anomalous in comparison with historical trends. Data from the 9-meter temperature sensor at the CKT136, KY site were suspect for 6 to 12 months. Amec Foster Wheeler reviewed data from the site and determined that six months of data need to be invalidated. The remaining six months can be rescaled. Amec Foster Wheeler received final results for analyses submitted for proficiency test (PT) study 0107 for Rain and Soft Waters from the National Laboratory of Environmental Testing (NLET), a branch of the National Water Research Institute (NWRI) with Environment Canada that provides quality assurance services. One ammonia sample was flagged “warning high.” Ammonia and sulfate results were “biased high.” These ratings require no formal corrective actions under the requirements of accreditation1. All other results were rated as “ideal.” Amec Foster Wheeler’s overall laboratory rating remains “very good.” During first quarter 2016, two subcontractors were subject to corrective action for their failure to properly verify operation of site equipment (erroneous entry of temperature calibration factors at CKT136, KY and failure to maintain specified operating temperatures during ozone instrument reverification at CVL151, MS). Table 1 lists the quarters of data that were validated to Level 3 during first quarter 2016 by site calibration group. Table 2 lists the sites in each calibration group along with the calibration schedule. Table 3 presents the measurement criteria for laboratory filter pack measurements. These criteria apply to the QC samples listed in the following section of this report. Table 4
1 Formal corrective actions are required for: Individual test results that are greater than 3σ from the assigned value. Youden average rank for a parameter outside of the 95 percent confidence interval of the overall rank with a
bias percent slope greater than an absolute value of 5. Consecutive study results for the same parameter with Youden average rank outside of the 95 percent
confidence interval of the overall rank without regard to bias percent slope. Three or more parameters with Youden average rank outside of the 95 percent confidence interval of the overall
rank in a single study without regard to bias percent slope.
presents the critical criteria for ozone monitoring. Table 5 presents the critical criteria for trace-level gas monitoring. Quality Control Analysis Count The QC sample statistics presented in this report are for reference standards (RF) and continuing calibration verification spikes (CCV) used to assess accuracy and for replicate sample analyses (RP) used to assess “in-run” precision. In addition, laboratory method blanks (MB) containing reagents without a filter; laboratory blanks (LB) containing reagents and a new, unexposed filter; and field blanks (FB) containing reagents and an unexposed filter that was loaded into a filter pack assembly and shipped to and from the monitoring site while remaining in sealed packaging are also included. Table 6 presents the number of analyses in each category that were performed during first quarter 2016. Sample Receipt Statistics Ninety-five percent of field samples from EPA-sponsored sites must be received by the CASTNET laboratory in Gainesville, FL no later than 14 days after removal from the sampling tower. Table 7 presents the relevant sample receipt statistics for first quarter 2016. Data Quality Indicator (DQI) Results Figures 1 through 3 present the results of RF, CCV, and RP QC sample analyses for first quarter 2016. All results were within the criteria listed in Table 32. Table 8 presents summary statistics of critical criteria measurements at ozone sites collected during first quarter 2016. The statistics presented contain data validated at Level 2 and Level 3. All data associated with QC checks that fail to meet the criteria listed in Table 4 were or will be invalidated unless the cause of failure has no affect on ambient data collection, and passing results still meet frequency criteria. Results in shaded cells either exceed documented criteria or are otherwise notable. Table 9 presents observations associated with the shaded cell results in Table 8. Table 10 presents summary statistics of critical criteria measurements at trace-level gas monitoring sites collected during first quarter 2016. The statistics presented contain data validated at Level 2 and Level 3. All data associated with QC checks that fail to meet the criteria listed in Table 5 were or will be invalidated unless the cause of failure has no affect on ambient data collection, and passing results still meet frequency criteria. Results in shaded cells either exceed documented criteria or are otherwise notable. Table 11 presents observations associated with the shaded cell results in Table 10. 2 The highest value depicted in Figure 3 is 20.4 percent. This is within criteria per the established rounding rules.
Laboratory Control Sample Analysis The laboratory control sample (LCS) is a reagent blank spiked with the target analytes from the established analytical methods and carried through the same extraction process that field samples must undergo. The LCS is not required by the CASTNET QA/QC program. LCS analyses are performed by the laboratory to monitor for potential sample handling artifacts and provide a means to identify possible analyte loss from extraction to extraction. Figure 4 presents LCS analysis results for first quarter 2016. All recovery values were between 85 percent and 120 percent. Blank Results Figures 5 through 7 present the results of MB, LB, and FB QC sample analyses for first quarter 2016. All first quarter results were within criteria (two times the reporting limit) listed in Table 3 with the exception of one Teflon FB sample with potassium at 16 times the reporting limit. All other QC results associated with this sample were within criteria. Sample results from THR422, ND, the associated site, were within the expected range for the site and season. Suspect/Invalid Filter Pack Samples Filter pack samples that were flagged as suspect or invalid during first quarter 2016 are listed in Table 12. This table also includes associated site identification and a brief description of the reason the sample was flagged. During first quarter, 21 filter pack samples were invalidated. Field Problem Count Table 13 presents counts of field problems affecting continuous data collection for more than one day for first quarter 2016. The problem counts are sorted by a 30-, 60-, or 90-day time period to resolution. A category for unresolved problems is also included. Time to resolution indicates the period taken to implement corrective action. References Amec Foster Wheeler Environment & Infrastructure, Inc. (Amec Foster Wheeler). 2014. Clean
Air Status and Trends Network (CASTNET) Quality Assurance Project Plan (QAPP) Revision 8.2. Prepared for U.S. Environmental Protection Agency (EPA), Office of Air and Radiation, Clean Air Markets Division, Washington, DC. Contract No. EP-W-15-003. Gainesville, FL. https://java.epa.gov/castnet/documents.do.
American Society for Testing and Materials (ASTM). 2008. ASTM E29-08, Standard Practice
for Using Significant Digits in Test Data to Determine Conformance with Specifications. ASTM International, West Conshohocken, PA, DOI:10.1520/E0029-08. www.astm.org.
U.S. Environmental Protection Agency (EPA). 2015. Appendix A to Part 58 – Quality Assurance Requirements for State and Local Air Monitoring Stations (SLAMS), Special Purpose Monitors (SPMs), and Prevention of Significant Deterioration (PSD) Air Monitoring. 40 CFR Part 58.
Table 1 Data Validated to Level 3 during First Quarter 2016 Calibration
Group* Months
Available Number of
Months Complete Quarters
Number of Quarters
E-3/W-10† May 2015 – October 2015
6 Quarter 3 2015 1
SE-4/MW-6‡ July 2015 – December 2015
6 Quarter 3 2015 – Quarter 4 2015
2
Notes: * The sites contained in each calibration group are listed in Table 2. † Contains ROM206 of the ROM406/ROM206 collocated pair ‡ Contains MCK131/231 collocated pair
Table 2 Field Calibration Schedule for 2016
Calibration Group
Months Calibrated
Sites Calibrated
Eastern Sites (24 Total) E-1
(8 Sites) February/August BEL116, MD WSP144, NJ ARE 128, PA PED108, VA
BWR139, MD CTH110, NY PSU106, PA VPI120, VA E-2
(11 Sites) April/October ABT147, CT WST109, NH NIC001, NY EGB181, ON
ASH135, ME CAT175, NY WFM007, NY UND002, VT HOW191, ME HWF187, NY WFM105, NY
E-3 (5 Sites)
May/November KEF112, PA LRL117, PA CDR119, WV MKG113, PA PAR107, WV
4 ) IC 20 95 – 105 0.040 1.0 Notes: 1 This column lists precision goals for both network precision calculated from collocated filter samples and laboratory precision based on
replicate samples. 2 This column lists laboratory accuracy goals based on reference standards and continuing calibration verification spikes. The criterion is
90–110 percent for ICP-OES reference standards. AC = automated colorimetry IC = ion chromatography ICP-OES = inductively coupled plasma-optical emission spectrometry MARPD = mean absolute relative percent difference mg/L = milligrams per liter µg/Filter = micrograms per filter * = as nitrogen Values are rounded according to American Society for Testing and Materials (ASTM) E29-08, Standard Practice for Using Significant
Digits in Test Data to Determine Conformance with Specifications (ASTM, 2008). For more information on analytical methods and associated precision and accuracy criteria, see the CASTNET QAPP, (Amec Foster
Wheeler, 2014).
Table 4 Ozone Critical Criteria*
Type of Check Analyzer Response Zero Less than ± 3 parts per billion (ppb)
Span Less than or equal to ± 7 percent between supplied and observed concentrations
Single Point QC Less than or equal to ± 7 percent between supplied and observed concentrations Notes: * Applies to CASTNET sites that are configured and operated in accordance with Part 58 of Title 40 of the Code of Federal Regulations
(EPA, 2015). The minimum frequency for these checks is once every two weeks.
Values are rounded according to ASTM E29-08, Standard Practice for Using Significant Digits in Test Data to Determine Conformance with Specifications (ASTM, 2008).
Table 5 Trace-level Gas Monitoring Critical Criteria*
Parameter Analyzer Response
Zero Check Span Check / Single Point QC Check
SO 2 Less than ± 3 ppb
Less than or equal to ± 10 percent between supplied and observed concentrations NOy Less than ± 3 ppb
CO Less than ± 40 ppb
Notes: *Applies to CASTNET sites that are configured and operated in accordance with Part 58 of Title 40 of the Code of Federal Regulations (EPA, 2015). The minimum frequency for these checks is once every two weeks.
Values are rounded according to ASTM E29-08, Standard Practice for Using Significant Digits in Test Data to Determine Conformance with Specifications E29 (ASTM, 2008).
SO2 = sulfur dioxide NOy = total reactive oxides of nitrogen CO = carbon monoxide ppb = parts per billion
Notes: 1 Percentage of comparisons that pass the criteria listed in Table 4. Values falling below 90 percent are addressed in Table 9. 2 Absolute value of the average percent differences between the on-site transfer standard and the site monitor. Values exceeding the
criteria listed in Table 4 are addressed in Table 9. 3 90 percent confidence limit of the coefficient of variation. This should be less than or equal to the 7 percent single point QC check
critical criterion. Values exceeding this criterion are addressed in Table 9. %D = percent difference CL = confidence limit ppb = parts per billion
ROM206, CO NOy 100.00 1.34 100.00 2.48 0.27 100.00 0.44
Notes: 1 Percentage of comparisons that pass the criteria listed in Table 5. Values falling below 90 percent are addressed in Table 11. 2 Absolute value of the average percent differences between the supplied and observed concentrations. Values exceeding the criteria
listed in Table 5 are addressed in Table 11. 3 90 percent confidence limit of the coefficient of variation. This should be less than or equal to the 10 percent single point QC check
critical criterion. Values exceeding this criterion are addressed in Table 11. %D = percent difference CL = confidence limit ppb = parts per billion Table 11 Trace-level Gas QC Observations for First Quarter 2016
Site ID Parameter QC Criterion Comments BEL116, MD SO2 % Span Pass
Span |%D | % Single Point QC Pass Single Point QC |%D|
The analyzer malfunctioned 3/11/16. Repairs were completed and calibration verified on 3/23/16. Associated data were invalidated.
BEL116, MD NOy % Span Pass % Single Point QC Pass Zero Average
The analyzer drifted out of calibration 1/7/16 and was recalibrated on 1/21/16. Associated data were invalidated.
BVL130, IL CO % Single Point QC Pass The analyzer drifted out of calibration 2/28/16 and was recalibrated on 3/18/16. Associated data were invalidated.
Table 12 Filter Packs Flagged as Suspect or Invalid during First Quarter 2016
Site ID Sample No. Reason
BEL116, MD 1609001-06 Insufficient flow data due to problems with telemetry BFT142, NC 1605001-07 Insufficient flow volume CHE185, OK 1611004-02 Insufficient flow data due to problems with telemetry COW005, NC 1602003-01 Insufficient flow volume CTH110, NY 1602001-24 Insufficient flow volume due to power failure DCP114, OH 1611001-21
1612001-21 Insufficient flow data due to problems with telemetry
DIN431, UT 1602001-28 1604003-06 1605003-06
Insufficient flow volume due to a leak in the flow system
GAS153, GA 1602001-32 1608001-24
“Calibrator onsite” flags applied in error for 1602001-32. Flow data should be recoverable for this sample.
Insufficient flow data due to problems with telemetry for sample 1608001-24.
NIC001, NY 1602001-53 Insufficient flow volume due to power failure PRK134, WI 1612001-43 Insufficient flow volume ROM406, CO 1602001-67 Insufficient flow volume due to a leak in the flow system SPD111, TN 1603001-49
1611001-49 Insufficient flow volume for both samples. Problems with
telemetry for sample 1611001-49. WNC429, SD 1604003-23 Insufficient flow volume YOS404, CA 1604003-25
1609003-25 1611003-25
Insufficient flow volume due to a malfunctioning flow pump.